Fréchet Denoised Distance: Enhancing Plausibility Evaluation for Generated Designs with Denoising Autoencoder
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916465309908992 |
|---|---|
| author | Fan, Jiajie Trigui, Amal Bäck, Thomas Wang, Hao |
| author_facet | Fan, Jiajie Trigui, Amal Bäck, Thomas Wang, Hao |
| contents | A great interest has arisen in using Deep Generative Models (DGM) for generative design. When assessing the quality of the generated designs, human designers focus more on structural plausibility, e.g., no missing component, rather than visual artifacts, e.g., noises or blurriness. Meanwhile, commonly used metrics such as Fréchet Inception Distance (FID) may not evaluate accurately because they are sensitive to visual artifacts and tolerant to semantic errors. As such, FID might not be suitable to assess the performance of DGMs for a generative design task. In this work, we propose to encode the to-be-evaluated images with a Denoising Autoencoder (DAE) and measure the distribution distance in the resulting latent space. Hereby, we design a novel metric Fréchet Denoised Distance (FDD). We experimentally test our FDD, FID and other state-of-the-art metrics on multiple datasets, e.g., BIKED, Seeing3DChairs, FFHQ and ImageNet. Our FDD can effectively detect implausible structures and is more consistent with structural inspections by human experts. Our source code is publicly available at https://github.com/jiajie96/FDD_pytorch. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_05352 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Fréchet Denoised Distance: Enhancing Plausibility Evaluation for Generated Designs with Denoising Autoencoder Fan, Jiajie Trigui, Amal Bäck, Thomas Wang, Hao Computer Vision and Pattern Recognition A great interest has arisen in using Deep Generative Models (DGM) for generative design. When assessing the quality of the generated designs, human designers focus more on structural plausibility, e.g., no missing component, rather than visual artifacts, e.g., noises or blurriness. Meanwhile, commonly used metrics such as Fréchet Inception Distance (FID) may not evaluate accurately because they are sensitive to visual artifacts and tolerant to semantic errors. As such, FID might not be suitable to assess the performance of DGMs for a generative design task. In this work, we propose to encode the to-be-evaluated images with a Denoising Autoencoder (DAE) and measure the distribution distance in the resulting latent space. Hereby, we design a novel metric Fréchet Denoised Distance (FDD). We experimentally test our FDD, FID and other state-of-the-art metrics on multiple datasets, e.g., BIKED, Seeing3DChairs, FFHQ and ImageNet. Our FDD can effectively detect implausible structures and is more consistent with structural inspections by human experts. Our source code is publicly available at https://github.com/jiajie96/FDD_pytorch. |
| title | Fréchet Denoised Distance: Enhancing Plausibility Evaluation for Generated Designs with Denoising Autoencoder |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2403.05352 |